(1) T cell interactions with leukemia cells: We have found that T cells recognizing all the leukemia antigens so far tested occue in normal individuals, and that after SCT T cells specific for several antigens increase in what appears to be a permissive environment for antigen-specific T cell expansion in the first weeks after transplant. This finding underpins development of new trials currently awaiting regulatory approval involving vaccinating donors with PR1 and WT1 nonapeptides and boosting patients with the same vaccines early after SCT. Two clinical trials - a safety study where patients with myeloid malignancies received a single dose of PR1 and WT1 vaccine and a second study where they received multiple doses of vaccine over a 16 week period demonstrated efficacy in increasing T cell frequencies to either or both vaccines in all 8 patents. However multiple vaccination resulted in loss of T cell responses and a loss of high avidity T cells recognizing the leukemia peptides suggesting the induction of tolerance. Currently our clinical grade vaccines are restricted to HLA A2 individuals and only stimulate CD8 Tcells. We are screening a peptide library of WT1 to identify peptides which induce other HLA class 1 molecules and also CD4 T cells so as to design a more widely applicable and more potent vaccine. (2) Natural Killer (NK) cell GVL: In a retospective analysis of over 270 SCT we found that stem cell donors with the KIR genotype 2DS1, 3DS1, 2DL5a conferred significantly less relapse in recipients with acute myelogenous leukemia (AML). A subsequent study of functional and phenotypic differences in NK cells between donors with and without this "favorable" genotype has not so far identified any differences. (3) GVL effect of mismatched T cells: We have shown that alloresponding T cells exist in both the naive and memory T cell compartments. memory alloreactive T cells often show cross reactivity with DNA virus antigens of CMV and EBV viruses. Using T cell receptor gene sequencing we confirmed that individual T cell clones can show dual reactivity against viral antigens and HLA mismatched target cells. To study whether cross-reactivity might present a limitation to the use of third party adoptive transfer of virus specific T cell clones we studied cross reactivity of virus specific T cells given to patients with post SCT viral reactivation and showed that in vivo such cross-reactivity did not result in GVHD. We are now exploring whether mismatched alloreactive T cell lines can be generated using an artificial genetically modified K562 antigen-presenting cell to provide GVL effects and also whether alloreacting T cell lines generated primarily against DNA viruses also have GVL effects and could be used to treat relapsed leukemia after SCT. (4) Immune recovery after SCT : We measured plasma cytokine profiles and lymphocyte repertoire to determine the optimum time to boost graft-versus-leukemia (GVL) effects after transplant and to characterize immune recovery in different transplant protocols. We found two patterns of cytokine expression - an early surge of lymphoid and myeolid growth factors in the first few days after transplant and a later surge of lymphokines released around 4 weeks after transplant. We are now extending cytokine profiling to compare patterns in autologous and T cell replete SCT. We will correlate recovery of different lymphocyte subsets with individual cytokine profiles to determine the factors which boost optimum recovery. (5) Selective depletion of alloreactive T cells: A clinical trial to selectively deplete T cells from the SCT that can cause GVHD and further translational work to develop improved techniques to selectively photodeplete activated T cells using rhodamine-based dyes has now been completed. We evaluated the ability of this photodepletion technique to selectively deplete graft-versus-host disease (GVHD) alloreacting T cells from stem cell transplants. Donor lymphocytes were stimulated with irradiated in-vitro expanded recipient T lymphocytes. Alloactivated T cells preferentially retaining the photosensitizer 4, 5-dibromorhodamine 123 (TH9402) (Kiadis Pharma, NL) were eliminated by light exposure. Twenty-four patients with hematological malignancies (16 high-risk) conditioned with fludarabine, cyclophosphamide, and total body irradiation received a CD34-selected stem cell allograft from an HLA identical sibling and 5x106/kg selectively depleted (SD) donor T cells. Low-dose cyclosporine was used as the only post-transplant immunosuppression. The overall probability of grade III-IV acute-GVHD was 13%. Fourteen patients developed (predominantly limited) chronic GVHD (c-GVHD). Five patients relapsed, 2 of whom remain alive in remission after further treatment. Thirteen patients survive at a median of 22 months. Overall survival and disease free survival actuarial probabilities (+ SEM) were 43 + 13% and 35 + 13% respectively. The SD technique resulted in a low incidence of relapse (24 + 10%), but was complicated by late non-relapse mortality associated with c-GVHD and infection of 50 + 14% at 4 years follow up. These results suggest that SD may effectively reduce severe a-GVHD while conserving graft-versus leukemia effects. (6) Mesenchymal stromal cells (MSC): Preclinical studies exploring the interaction of MSC with cytotoxic T cells showed that MSC can reduce cytotoxicity of T effector cells against leukemia lines. Further studies are warranted to explore a possible suppressive effect of MSC infusions on GVL reactivity. We have worked with the Clinical Center Cell Processing Section to develop clinical grade MSC for three trials covering the use of MSC to treat GVHD , poor marrow function and major organ damage. (7) A Clinical trial of T cell depleted SCT as a platform for immunotherapeutic approaches is ongoing. Thirty-six patients with hematologic malignancies underwent allogeneic HSCT from their HLA-identical siblings. The median age was 43 years (range 16-68). Transplant indications were acute leukemia (24), other myeloid malignancy (9), other lymphoid malignancy (3). Subjects received myeloablative conditioning regimen with cyclophosphamide, fludarabine and total body irradiation (12 Gy with lung shielding to 6 Gy). Subjects 55 years of age and older received 4 Gy. G-CSF mobilized peripheral blood stem cells from the donor were CD34+ selected utilizing the Miltenyi CliniMacs system, with infusion of a target CD34+ dose of 6 x 106/kg (range 3 to 10 x 106/kg) and a fixed CD3+ dose of 5 x 104 /kg. Patients received low-dose cyclosporine alone till day 21. Delayed lymphocyte add back (5 x 106 CD3+/kg) was given on day +90. Day 200 overall survival was 79%. Thirty-four subjects achieved complete donor (>95%) myeloid chimerism by day 14 and the median time to complete donor CD3+ chimerism was 45 days. The incidence of chronic GVHD was 34.3%. At a median follow up of 3.6 years, Kaplan-Meier estimates of relapse, nonrelapse mortality and overall survival were 35%, 33% and 44% respectively. These results represent acceptable engraftment and clinical outcomes. The trial is ongoing. A trial of alemtuzemab in LGL leukemia is ongoing Patients had chronic cytopenia, increased clonal CD3+CD8+ CD16+ or CD57+ T-LGL. After 1 mg test dose, alemtuzumab was administered at 10 mg/d for 10 days. Overall nine (56%;95% CI 30%-80%) patients responded to alemtuzumab with five achieving a CR and four a PR. Of the 6 anemic patients who responded to therapy the median pre-treatment hemoglobin was 7.3 (range, 5.4-9.3) g/dL and the median post-treatment hemoglobin was 12.2 (range 10-13.7) g/dL. Three responders with an ANC <500 /cu mm pre-treatment improved to an ANC >1,000/cu mm after alemtuzumab. This study showed that alemtuzumab was safe and showed promising activity in pre-treated T-LGL patients.